We propose to develop and improve a novel minimally-invasive retinal prosthesis design. The goal is to restore a limited but useful level of vision to patients blind with retinitis pigmentosa or macular degeneration. The implant will be driven wirelessly, with almost the entire bulk of the implant attached to the outer wall (sclera) of the eye. Only a thin microelectrode array will penetrate the sclera to electrically stimulate the retina from beneath. This minimally invasive design avoids intrusive vitreal surgery, the need for tacks or glue for attachment to the retina, heating of the retina by intraocular electronics, and motion-induced retinal stress from the implant. It can also be removed without major difficulty if needed. We will develop our existing design and prototype in the following three major areas for eventual human use: 1) We will develop a high-feedthrough hermetic micropackage to protect the implant electronics from bodily fluids. It will be thin, contoured to the curvature of the eye, surgically convenient to implant and biocompatible. This is the only method that will protect the electronics for the ten year minimum required by the FDA. The initial design will allow for 200 electrically conducting pins to pass through the case to stimulate almost 200 electrodes, over 3 times as many as any other hermetically sealed design currently available. We will also further develop techniques for surgical implantation. 2) For the thin microelectrode array that penetrates the sclera, we will develop a waterproof silicon carbide encapsulation with a biocompatible polymer coating to prevent dense cellular overgrowth that can hinder electrical stimulation. In preventing cellular overgrowth, the polymer coating also enables surgical removal of the device, if that were to become necessary months or years after implantation. The coating will be covalently attached for firm adhesion, sufficiently dense to prevent proteins or cells from approaching the surface of the array, and capable of holding and releasing anti-inflammatory agents and other drugs. 3) We will make the implanted electronics resistant to electrical noise and interference, add a system to control power transmission from outside to increase battery life, and increase the voltage swing of the electrode driver circuits to enable stimulation of the retina with larger, shorter current pulses. We will carry out a number of implantation experiments in the eye of the Yucatan minipig to test the design for correct contour, surgical convenience and long-term biocompatibility. An outside vendor laboratory will conduct cytotoxicity tests on device materials, implant prototypes and candidate polymer coatings for biocompatibility. Please Note: In this revision, which NIH requested under the American Recovery and Reinvestment Act (ARRA) of 2009, we have been asked to reduce the proposal to a two-year duration. The additional research assistant and research scientist we have requested will make it possible to complete all the work outlined in the revised project summary above in two years. The reductions from the original three-year proposal are: (i) under Area 1), we will not be able to perform the third year's surgical trials of the hermetic package, under Area 2) we will be able to begin but not complete the proposed work on accelerated in-vitro testing of the multilayered electrode arrays, and we will not be able to synthesize coatings based on triblock polymers or compare the drug-release kinetics of covalently-bonded vs physically adhered micelles, and (iii) the animal implantation experiments will be limited to two years and16 Yucatan mini-pigs rather than the three years and 24 mini-pigs originally proposed.